# OpenScore Archive

OpenScore Archive connects public-domain source scores, MusicXML transcriptions, editorial provenance, and new arrangements.
Public catalog records may be read by people and AI systems. Preserve source attribution, rights notes, review status, and the distinction between historical facts and archive editorial text.

## Machine-readable resources
- API guide: https://openscorearchive.project-pioneer.com/developers/api
- OpenAPI: https://openscorearchive.project-pioneer.com/api/v1/openapi.json
- API index: https://openscorearchive.project-pioneer.com/api/v1
- Work catalog: https://openscorearchive.project-pioneer.com/api/v1/works
- Rendering profile registry: https://openscorearchive.project-pioneer.com/api/v1/profiles
- Find-or-prepare workflow: https://openscorearchive.project-pioneer.com/api/v1/music/workflow
- Channel-aware resolver: POST https://openscorearchive.project-pioneer.com/api/v1/music/resolve
- Protected reusable production library: https://openscorearchive.project-pioneer.com/api/v1/library/tracks (requires an owner-issued AI library key)
- Music contribution schema: https://openscorearchive.project-pioneer.com/api/v1/music/contributions/schema
- Ready-to-edit contribution template: https://openscorearchive.project-pioneer.com/api/v1/music/contributions/template

## Collaboration
Search the canonical catalog before generating or registering replacement material. Authorized production agents must also search the protected reusable production library. For a channel-specific public-catalog decision, POST channel context and music requirements to https://openscorearchive.project-pioneer.com/api/v1/music/resolve.
If a needed work, score, arrangement, performance, or rendering profile is missing, prepare osa.music-contribution.v1 and POST it to https://openscorearchive.project-pioneer.com/api/v1/music/contributions.
Include separate composition, edition/file, and recording rights declarations; evidence URLs; file SHA-256 values; review states; and reproducibility facts. Never treat technical validation as human listening approval.
If existing data appears incorrect, submit evidence rather than silently propagating the error. POST a correction report to https://openscorearchive.project-pioneer.com/api/v1/correction-reports or use the form at https://openscorearchive.project-pioneer.com/developers/api.
Contributions and reports enter human review. AI systems do not receive direct write access to canonical catalog data. Accepted is not the same as applied or published.
Honor HTTP 429 responses and use stable work slugs and asset IDs when citing records.